RDS App v1.0.0

RDS App v1.0.0

Initial public release of the Respondent-Driven Sampling (RDS) Application — an open-source web app for conducting RDS surveys of unsheltered populations during Point-in-Time (PIT) counts.

Developed by the University of Washington (Department of Sociology, iSchool) and the UW Scientific Software Engineering Center (SSEC) in collaboration with the King County Regional Homelessness Authority (KCRHA). The first production deployment supported the 2026 King County Unsheltered PIT Count, collecting surveys from approximately 2,183 individuals.

Highlights

• Volunteer survey collection — Volunteers authenticate via SMS OTP, scan a referral QR code, and guide participants through the survey on a mobile-friendly interface.
• Referral-chain tracking — Each completed survey generates three child QR codes, preserving the RDS network structure in the database.
• Role-based access control — Four roles (volunteer, manager, admin, super-admin) with attribute-based permissions via CASL.
• Admin dashboard — View, filter, and manage all survey entries and volunteer accounts.
• CLI tooling — Scripts for creating super-admin accounts, importing survey locations, and generating seed/coupon QR-code PDFs.
• SMS gift-card distribution — Optional bulk SMS to participants via Twilio.
• Infrastructure-as-code — Pulumi (TypeScript) configuration for Azure App Service deployment.
• Comprehensive documentation — MkDocs site covering setup, deployment, survey operations, CLI reference, and post-survey analysis.

Tech Stack

React + TypeScript (Vite) | Node.js/Express | MongoDB (Mongoose) | Twilio Verify | Azure App Service

Reuse

Licensed under BSD 3-Clause. The survey questionnaire, locations, and campaign content are all configurable without code changes — designed for adoption by other cities and localities.

Links

• Documentation: https://ssdlab.github.io/respondent-driven-sampling/
• Repository: https://github.com/SSDALab/respondent-driven-sampling